seo-technical
Pass
Audited by Gen Agent Trust Hub on May 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute local Python scripts using the shell, passing a user-supplied URL as a command-line argument. If the URL is not properly validated by the agent or the script, it could lead to command injection.
- Evidence:
python scripts/pagespeed_check.py <url> --json(SKILL.md) - Evidence:
python scripts/crux_history.py <url> --json(SKILL.md) - Evidence:
python scripts/gsc_inspect.py <url> --json(SKILL.md) - [PROMPT_INJECTION]: The skill is designed to process external data retrieved from target websites (meta tags, robots.txt, sitemaps, and JavaScript content), which represents a classic surface for indirect prompt injection.
- Ingestion points: Scraped content, meta tags, and robots.txt files from the target audit URL (SKILL.md).
- Boundary markers: No explicit delimiters or instructions are provided to the agent to treat external content as untrusted data rather than instructions.
- Capability inventory: The skill can write to the local file system (
.seo-cache/) and execute subprocesses (python scripts/...). - Sanitization: No evidence of validation or sanitization of retrieved web content before processing is mentioned.
Audit Metadata